The Delegate Fight: 2016

[Mr. Morden]

In 2012, there was a beauty contest, but it was ignored. In 2016, it will be used to pledge the statewide delegates. It appears that the delegate ballot will be down-ballot following senatorial and congressional nominations. Some voters will likely skip the delegate ballot.

If stereotypes are valid, this may hurt Trump. The Cruz campaign has probably done a better job of targeting supporters, and will make sure they understand the ballot. If Trump relies more on a mass appeal, it may be difficult to communicate the nuances.

Do we have any numbers from either 2008 or 2012 as to how many voters voted in the beauty contest but didn't vote in the delegate election (or didn't vote for a full slate of delegates)?

[Joe Biden 2020]

I know this was probably mentioned earlier in the thread, but I don't want to dig.  Are all the GOP contests tomorrow WTA?

[Gass3268]

I know this was probably mentioned earlier in the thread, but I don't want to dig.  Are all the GOP contests tomorrow WTA?

Florida and Ohio: WTA
Illinois: 15 statewide delegates are WTA, 3 delegates are directly elected by each congressional district (total of 54)
Missouri: WTA if over 50%, 12 statewide delegates are WTA, 5 WTA delegates by congressional district (total of 40)
North Carolina: Proportional, no threshold

[Joe Biden 2020]

I know this was probably mentioned earlier in the thread, but I don't want to dig.  Are all the GOP contests tomorrow WTA?

Florida and Ohio: WTA
Illinois: 15 statewide delegates are WTA, 3 delegates are directly elected by each congressional district (total of 54)
Missouri: WTA if over 50%, 12 statewide delegates are WTA, 5 WTA delegates by congressional district (total of 40)
North Carolina: Proportional, no threshold

Ahh, okay, thanks.  I knew FL and OH were.  I wasn't sure about the other three.  What about Arizona and Utah on March 22?

[Gass3268]

I know this was probably mentioned earlier in the thread, but I don't want to dig.  Are all the GOP contests tomorrow WTA?

Florida and Ohio: WTA
Illinois: 15 statewide delegates are WTA, 3 delegates are directly elected by each congressional district (total of 54)
Missouri: WTA if over 50%, 12 statewide delegates are WTA, 5 WTA delegates by congressional district (total of 40)
North Carolina: Proportional, no threshold

Ahh, okay, thanks.  I knew FL and OH were.  I wasn't sure about the other three.  What about Arizona and Utah on March 22?

Arizona: WTA
Utah: Proportional, 15% threshold, WTA if majority (Cruz will probably get a majority)

[Erc]

Per discussions here, I'm calling the remaining 14 delegates up for grabs in Wyoming for Cruz.

This could change at the state convention, but a Cruz sweep here seems overwhelmingly likely given the results of the county conventions.

[jimrtex]

In 2012, there was a beauty contest, but it was ignored. In 2016, it will be used to pledge the statewide delegates. It appears that the delegate ballot will be down-ballot following senatorial and congressional nominations. Some voters will likely skip the delegate ballot.

If stereotypes are valid, this may hurt Trump. The Cruz campaign has probably done a better job of targeting supporters, and will make sure they understand the ballot. If Trump relies more on a mass appeal, it may be difficult to communicate the nuances.

Do we have any numbers from either 2008 or 2012 as to how many voters voted in the beauty contest but didn't vote in the delegate election (or didn't vote for a full slate of delegates)?

The Illinois State Board of Elections has the results, but doesn't include information about undervotes, to make a definitive answer.

933,454 votes were cast in the presidential beauty contest. Since it was top of the ballot, voters would be likely to vote in that race even if they understood that it didn't matter. Presumably, there would be emphasis placed on being sure to vote in the delegate selection area, and to vote for the proper number of delegates candidates. That is a fairly complicated message, without also telling  voters that their first vote doesn't matter.

Compare - Top of the ballot:

**********************************
OR PRESIDENT OF THE UNITED STATES
(Vote for one)

( ) TED CRUZ
**********************************

Vs. Down ballot (below US Senator, Comptroller, and US Representative)

**********************************
FOR DELEGATE TO THE
NATIONAL NOMINATING CONVENTION
SIXTEENTH CONGRESSIONAL DISTRICT
(PLEASE NOTE:  Next to the name of each candidate for delegate appears
in parentheses the candidate's preference for President of the United States
or the word "uncommitted".)
(Vote for not more than three)

( ) TIM BIVINS (CRUZ)
***********************************************

In 2012, the number of delegates per district varied from 2 to 4 (two districts had two delegates, two districts had four, and all the rest had three). So I calculated the number of votes for Republican delegate candidates in each district, divided by the number of delegate positions for the district. This provides an estimate of the number of voters who voted in the district, assuming they had voted for the maximum number of candidates.

Statewide, this gives an estimate of 810, 986 voters, or 86.9% of those who voted in the beauty contest.

We can do the same comparison for the individual candidates:

Romney: 88.5%
Paul: 102.1%
Gingrich: 116.4%
Santorum: 75.9%
Perry: 4.4% (one delegate candidate)
Roemer: 8.9% (delegate candidate slate in only one district)

Romney, Paul, and Gingrich had complete slates. Santorum did not have slates in 4 of 18 districts, but they were relatively small, with perhaps 10% of the vote. It is reasonable to assume that many of his supporters voted for Gingrich delegates, since by that point it was mainly a Romney-Santorum contest. It would be reasonable for Santorum campaigner to suggest voting for Gingrich delegates in those districts.

The beauty contest was Romney 46.7%, Santorum 35.0%, Paul 9.3%, Gingrich 8.0%, Perry 0.6%, and Roemer 0.4%.

In IL-13 the delegate vote share was Romney 52.3%, Paul 20.5%, Gingrich 27.2%
In IL-7 Romney 66.4%, Paul 18.3%, Gingrich 15.4%
In IL-5 Romney 64.6%, Paul 19.6%, Gingrich 15.7%
In IL-4 Romney 56.0%, Paul 24.3%, Gingrich 19.7%

So all candidates overperformed in these areas. Note that IL-13 had about half the total Republican votes. It is a competitive district (part of the Illinois Democratic dummymander) including parts of Springfield, and thus likely to have an appreciable Republican presence in the primary. In the other 3 districts, there may have been little effort to reach Republican voters, and Romney delegates may have picked up more of the Santorum support.

If we compare the total vote for Santorum-Paul-Gingrich delegate candidates vs the three candidates in the beauty contest we get 85.1%, slightly lower than Romney, and likely reflecting leakage from Santorum supporters in the four districts.

I also looked at the sum of the votes for the 1st, 2nd, 3rd, and 4th candidates for each district. They vary, but in no consistent pattern. If voters were voting for one delegate, it was not necessarily for the first one. There is some personal voting for individual candidates:

Dan Rutherford (Illinois Treasurer) received 20% more than the other Romney delegates.
Ethan Hastert (former speaker Dennis Hastert's son) received 11% more than the other Romney delegates.
Darrin LaHood (former Ray LaHood's son) received 60% more than the other Gingrich delegates. Darrin LaHood, since a 2015 special election is now the representative for the district.
Richard Grabowski who was also seeking a congressional nomination received 20% more than the other Santorum delegates.

We can also compare the delegate vote to the congressional nomination vote. It appears that there is a fairly consistent relationship in votes cast:

(1) Presidential beauty contest;

(2) Contested congressional nomination where Republican nominee was competitive (IL-12, IL-13, IL-16). All three districts are now Republican held.

(3T) Contested congressional nomination where Republican nominee has little hope (IL-1, IL-2, IL-3).
(3T) Delegate.

(5) Uncontested congressional nomination. An incumbent will do a little better.

So let's take the typical district with 3 delegates. We'll assume that every voter voted in the beauty contest. Most. but not all would vote for either three delegates, or one delegate, but not two (assumption by me).

If none voted for one delegate then it would be about:

100% beauty contest.
88% 3 delegates.
0% 1 delegate.

To make up for a 1% drop in those voting for three delegates we have to add 3% to the one delegate districts. For example:

100% beauty contest.
87% 3 delegates.
3% 1 delegate.

The latter two categories can not exceed 100% combined so the upper limit is:

100% beauty contest.
82% 3 delegates
18% 1 delegate.

If we assume that a voter who votes for delegates votes for three delegates 90% of the time, then the best solution is:

********************
100% beauty contest.

85% 3 delegates
9% 1 delegate
6% 0 delegates.
********************

Factors to consider tomorrow. There is a wide variation in Republican voters per district (roughly 10:1, 84K v 8K). All districts choose 3 delegates in 2016.

A typical Republican voter in a heavily Democratic district in Chicago is not a typical voter.  If it is a black district, they might be among the few black Republicans. If they are white, they might be living in a safe highrise (this might be an advantage for Kasich or Rubio). But Kasich doesn't have a full set of delegates. In IL-4 (the Hispanic, earmuffs district) there might be some residual white working class Republicans, who might vote for Trump. Since all Puerto Ricans are US citizens, this might be an advantage for Rubio or Cruz, except most voters are Democrats.

It might be easier to get voters to vote in both the presidential preference poll and the delegate selection, since it easier to say "be sure to vote for both, both determine delegates" and say it with conviction. In 2012 you might have to say "it is important to vote for both", but if you started explaining why it gets confusing.

[Erc]

In D.C., all the elected delegates were on the Rubio or Kasich slates; don't expect any surprises here at the convention.  Of course, what their second choices would be is another matter.

[Mr. Morden]

A typical Republican voter in a heavily Democratic district in Chicago is not a typical voter.  If it is a black district, they might be among the few black Republicans. If they are white, they might be living in a safe highrise (this might be an advantage for Kasich or Rubio). But Kasich doesn't have a full set of delegates. In IL-4 (the Hispanic, earmuffs district) there might be some residual white working class Republicans, who might vote for Trump. Since all Puerto Ricans are US citizens, this might be an advantage for Rubio or Cruz, except most voters are Democrats.

Where is Kasich missing delegates?  Looking at this list, I see 3 Kasich delegate candidates per district in every district.

[jimrtex]

A typical Republican voter in a heavily Democratic district in Chicago is not a typical voter.  If it is a black district, they might be among the few black Republicans. If they are white, they might be living in a safe highrise (this might be an advantage for Kasich or Rubio). But Kasich doesn't have a full set of delegates. In IL-4 (the Hispanic, earmuffs district) there might be some residual white working class Republicans, who might vote for Trump. Since all Puerto Ricans are US citizens, this might be an advantage for Rubio or Cruz, except most voters are Democrats.

Where is Kasich missing delegates?  Looking at this list, I see 3 Kasich delegate candidates per district in every district.

I misread an article that said he didn't have enough signatures. In Illinois, a petition is presumed valid unless someone challenges it. He allegedly did not have enough signatures in six districts.

[jimrtex]

A typical Republican voter in a heavily Democratic district in Chicago is not a typical voter.  If it is a black district, they might be among the few black Republicans. If they are white, they might be living in a safe highrise (this might be an advantage for Kasich or Rubio). But Kasich doesn't have a full set of delegates. In IL-4 (the Hispanic, earmuffs district) there might be some residual white working class Republicans, who might vote for Trump. Since all Puerto Ricans are US citizens, this might be an advantage for Rubio or Cruz, except most voters are Democrats.

Where is Kasich missing delegates?  Looking at this list, I see 3 Kasich delegate candidates per district in every district.

I realized that you could get a better estimate by looking at counties that are contained  within a single congressional district, since the beauty contest and the delegate contests cover the same areas. Unfortunately, Illinois CD's chop counties a lot.

LaSalle County (just outside the Chicago Metro) is entirely in IL-16. The delegate vote (divided by three) was 81.5% of the beauty contest vote. The alternate delegate vote count was 79.0%.

[Erc]

Trump wins the first truly WTA contest of the cycle, giving him all 9 delegates out of the Northern Marianas.  This is also his 8th state in which he has won a majority of the delegates, so Trump now qualifies to have his name placed into nomination at the convention under Rule 40.

[Erc]

March 15 Results

Trump wins Florida and its 99 delegates.

Kasich wins Ohio and its 66 delegates.  This is his first Rule 40 win, though with a Rubio dropout a D.C. Rule 40 win seems plausible as well.

Rubio suspends his campaign.  This releases some of his delegates automatically.  I'm not going to change anything on the main page for a while (apart from AK, possibly).  I'll refer you to this post for discussion of what happens to his delegates.

What should I do with Rubio's column on the main post?  Keep it around (there are a lot of delegates there); fold it into the Uncommitted/Other column?  Move Carson/Bush/etc into an "Other" column along with Rubio?

[Ljube]

After the poor results tonight, TRUMP absolutely needs all 95 delegates from New York. He must win more than 50% of the vote there.

[Erc]

After the poor results tonight, TRUMP absolutely needs all 95 delegates from New York. He must win more than 50% of the vote there.

Honestly, this night was about par for Trump.  That said, par for Trump is still, by my really dumb projections, still gonna end up around 40 delegates short at this pace.

That said, it wouldn't take much for him to do so (or come so close he certainly wins on the first ballot).  Winning in Indiana or Maryland would do it, or even Nebraska or South Dakota.

That said, there are also many states that might cripple him if he loses: California is the big one, obviously, but a loss in Wisconsin would be devastating, and being shut out of Pennsylvania would hurt big-time.

There's then the open question of, if he's short by 40-60 delegates, is that really enough to stop him?

[pppolitics]

After the poor results tonight, TRUMP absolutely needs all 95 delegates from New York. He must win more than 50% of the vote there.

New York is his 2nd best state after West Virginia.

[Erc]

Ides of March Democratic Results

Clinton 393 - Sanders 298

These are very preliminary results, based on shoddy or non-existent CD breakdowns.  In particular, I imagine Sanders' current delegate lead in IL will disappear once Chicago results come in.

Clinton now has a 318-delegate lead in pledged delegates.  Sanders would need to win 57% of all remaining pledged delegates to tie in pledged delegates (let alone the supers).

Especially given the results last night, I don't see Sanders cracking 57% of pledged delegates outside of the remaining few caucus states.

If Sanders is going for a minority platform at the convention, he'll almost certainly have it by April 26.  I will continue tracking the Democratic side until at least then.

[Erc]

Rubio's Delegates (Redux)

Considering just the first ballot here...

Iowa (7) - remain bound "regardless."  This also applies to Carson's 3 delegates and 1 delegate each for Bush, Paul, Fiorina, and Huckabee.

New Hampshire (2) - bound until Rubio "withdraws."  This also applies to Bush's 3 delegates.

Nevada (7) - as he has "suspended," they are released or reallocated (Rubio's option). This also applies to Carson's 2 delegates.

Alabama (1) - released automatically.

Alaska (5) - if Rubio has "dropped out," (viz. does not "maintain an active campaign"), his delegates are reallocated, 3 to Cruz and 2 to Trump.

Arkansas (9) - bound until released.

Georgia (16) - bound until Rubio "withdraws."

Massachusetts [8] - bound until released.

Minnesota (17) - bound until released or Rubio "withdraws."  If Rubio un-withdraws, he'd get them back.

Oklahoma (12) - bound until Rubio "is for any reason no longer a candidate."

Tennessee (9) - presumably bound regardless, though perhaps released if Rubio "withdraws."

Texas (3) - bound until Rubio "withdraws" or releases them.

Virginia (16) - apparently still bound, but unclear.  This also includes Carson's 3 delegates here.

Kansas (6) - bound until released.  

Kentucky (7) - bound until released.  If they are released, they are reallocated on the basis of a secret vote of the whole delegation.

Louisiana (5) - automatically released.

Puerto Rico (23) - unknown.

Hawaii (1) - bound unless Rubio has "withdrawn."

DC (10) - bound unless Rubio "withdraws."

Wyoming (1) - unclear, but presumably unbound as Rubio is no longer "still in the race."

North Carolina (6) - apparently bound indefinitely.  This also includes Carson's 1 delegate here.

Louisiana (and, trivially, Alabama) is the only truly clear-cut case where he loses the binding of his delegates now.  Nevada seems very likely, and Alaska seems likely as well.  New Hampshire and Oklahoma are less clear.

The rest largely depend on whether Rubio is considered to have "withdrawn."  Presumably not, but let's keep an eye out on this.  Any election law experts able to weigh in on this?

[Ljube]

Let us now try to estimate TRUMP's delegates in the remaining states.
Or is it too early for such an exercise?

[Beefalow and the Consumer]

Let us now try to estimate TRUMP's delegates in the remaining states.
Or is it too early for such an exercise?

I can attempt, using polls, some of them really stale.  Here are states with solid Trump leads and winner take all/most allocations:

March 22:
  Arizona: Winner take all, 58
April 19:
  New York: Winner take most, 95
April 26:
  Maryland: Winner take most, 38
  Pennsylvania: Winner take most, 71 (54 unbound)
May 3:
  Indiana: Winner take most, 57
June 7:
  New Jersey: Winner take all, 51

This is 370 just from six states that will be very, very hard for Trump to lose.

Other states that Trump should have in the bag:

April 26:
  Connecticut: 28
  Rhode Island: 19
  West Virginia: 34
 
Total: 451, which Trump should win almost all of.

States where Trump may be vulnerable to an upset:

March 22:
  Utah: 40
April 5:
  Wisconsin: Winner take most, 42
April 26:
  Delaware: Winner take all, 16
May 10:
  Nebraska: Winner take all, 36
May 17:
  Oregon: 28
May 24:
  Washington: 44
June 7:
  California: Winner take all, 172
  Montana: Winner take all, 27
  New Mexico: 24
  South Dakota: Winner take all, 29

Vulnerable delegates: 458

[Erc]

Let us now try to estimate TRUMP's delegates in the remaining states.
Or is it too early for such an exercise?

I can attempt, using polls, some of them really stale.  Here are states with solid Trump leads and winner take all/most allocations:

March 22:
  Arizona: Winner take all, 58
April 19:
  New York: Winner take most, 95
April 26:
  Maryland: Winner take most, 38
  Pennsylvania: Winner take most, 71 (54 unbound)
May 3:
  Indiana: Winner take most, 57
June 7:
  New Jersey: Winner take all, 51

This is 370 just from six states that will be very, very hard for Trump to lose.

Other states that Trump should have in the bag:

April 26:
  Connecticut: 28
  Rhode Island: 19
  West Virginia: 34
  
Total: 451, which Trump should win almost all of.

States where Trump may be vulnerable to an upset:

March 22:
  Utah: 40
April 5:
  Wisconsin: Winner take most, 42
April 26:
  Delaware: Winner take all, 16
May 10:
  Nebraska: Winner take all, 36
May 17:
  Oregon: 28
May 24:
  Washington: 44
June 7:
  California: Winner take all, 172
  Montana: Winner take all, 27
  New Mexico: 24
  South Dakota: Winner take all, 29

Vulnerable delegates: 458

My personal take here.  Trump needs 548 delegates to secure the nomination on the first ballot, though falling a few short will probably be close enough.

Definite Trump Majorities (285, of which 240 seem assured)

Arizona (58) and New Jersey (51) seem absolutely secure in Trump delegate sweeps.

West Virginia (34) likewise seems secure, although a small handful may slip through due to loophole primary issues.  Also note that it's unclear whether West Virginia's delegates are actually bound.  Presumably, those who filed under Trump are actually Trump supporters, but there does not seem to be a vetting process in WV.

New York (95) has 50% WTA thresholds on the CD and the At-Large level.  Can't be entirely certain that he will hit that figure statewide, but 70 at least seems assured.  If he does hit it statewide, there's always the possibility he falls below it in some NYC-area CDs.

Connecticut (28) a win seems assured (along with presumably a win in at least 4 CDs).  CDs are WTA, statewide is proportional with a 50% WTA threshold.  He'll pick up at least 17 delegates here, with all 28 possible if he does very well.  Keep an eye out for CD 4, though, which may be prime Kasich territory.

Rhode Island (19) is proportional At-Large and by CD; this is likely to limit him to 9-10 delegates barring someone falling below the 10% threshold.

Unclear (405)

Pennsylvania (71) is a complete mess due to its loophole primary.  Kasich is obviously making a play for the state.  Trump win seems more likely than not, but whether that turns into any more than 17 delegates is another question.

Delaware (16) is WTA.  It's an odd state (see McDonnell, Christine); voted for McCain on Super Tuesday 2008 and Bush right after NH in 2000.  Trump seems favored, though maybe Kasich would have an outside shot if he campaigns here.  The more wacko elements of the DE GOP would seem to favor Trump over Cruz.

Maryland (38) is WTA by CD and At-Large.  MD seems like it would be a prime target for Kasich.  Trump isn't polling exceptionally well here, DC suburbs would be Kasich territory.  Whether they'd be enough to carry the state in a three-man race is unclear, though.

Indiana (57) is WTA At-Large and by CD.  Odds favor Trump in a divided field (see Michigan and Illinois) but perhaps Kasich could make a play here with Rubio gone.

Wisconsin (42) is WTA.  Odds seem to heavily favor Trump, unless Kasich steers well clear of the state.  Even then, it would be an uphill battle for Cruz.

American Samoa (9) has an unbound delegation.  Could go for Trump, but I have no idea.

California (172): the big kahuna, WTA At-Large and by CD.  I honestly have no clue, but I'd give the edge to Trump in a divided field.

Likely Not Trump Majority (293, of which Trump wins maybe 38)

Washington (44): Trump may be slightly disfavored, but the state is proportional so it doesn't much matter.

New Mexico (24): I don't know, but it's proportional.

Oregon (28): proportional.  Cruz seems favored, but Trump will still win some delegates.

Utah (40).  Only question is whether Cruz hits the 50% WTA threshold.

North Dakota (28).  Unbound, chosen by convention.  Trump has no chance.

Colorado (37): chosen by convention.  I would be highly surprised if Trump wins more than 3 delegates.

Nebraska (36): WTA.  Seems like Cruz country.

Montana (27): WTA for Cruz.

South Dakota (29): WTA for Cruz.

TL;DR:  Giving Trump wins in all the states I'd favor him in puts him around 15 delegates past a majority.  That said, a lot of that is contingent on WTA or almost-WTA wins.  Basically, everything would have to go right for him to win an outright majority.  Obviously, results are going to be correlated, but Trump will probably have a shock loss in at least one or two of those states, which would push the outcome out past June 7 (even if it's decided before Cleveland).

If he loses two or three of them, though, or one of them is California, then it's anyone's game and this thing is going to multiple ballots.

[Ljube]

Good analysis. Are we sure Montana is Cruz country?

EDIT: TRUMP is going to have to make a play for NE, SD and MT. He needs every delegate he can get. SD and MT are late enough that he may win them solely on his message of party unity.

EDIT2: Unpledged delegates, as well as released delegates should be considered as possible TRUMP supporters.

[Erc]

Good analysis. Are we sure Montana is Cruz country?

EDIT: TRUMP is going to have to make a play for NE, SD and MT. He needs every delegate he can get. SD and MT are late enough that he may win them solely on his message of party unity.

EDIT2: Unpledged delegates, as well as released delegates should be considered as possible TRUMP supporters.

All good points, though the unpledged/other delegates won't be resolved presumably before June 7, except perhaps in a very small number of cases.

As a historical comparison, Kasich's win in Ohio marks the first time four different candidates have won GOP contests in the same year in the modern primary era.  (Recall that Ron Paul did not win the Virgin Islands in 2012 by any reasonable definition, or limit yourself to the 50 states if you wish).

[muon2]

From what I see the AP has called the IL CDs where Cruz has delegates in the top three (IL-13, 15, 16, 17, 18) and he has 9 delegates out of those. The Kasich CDs are not completely called yet, but Kasich has four delegates that are called (2 each in IL-7 and 9). Kasich has two other delegates in third place but not called yet (IL-6 and 7). His two opportunities in IL-10 slipped away as the last precincts came in, though AP hasn't called the second and third places yet.

[The Other Castro]

I can't speak for all members of Fairfield County (CT-4), but we're definitely the type that are receptive to Kasich.